HONG KONG, Sept. 26, 2025 (GLOBE NEWSWIRE) — Kuaishou Technology (“Kuaishou” or the “Company”; HKD Counter Stock Code: 01024 / RMB Counter Stock Code: 81024), a leading content community and social platform, announced the launch of its Kling AI 2.5 Turbo Video Model on September 23. The latest model features major upgrades to its text-to-video and image-to-video capabilities, significantly enhancing generation quality with industry-leading performance, while further reducing utilization costs for greater affordability.
Industry-Leading Evaluation Results
In a blind test with professionals evaluating model performance, Kling AI’s 2.5 Turbo Video Model outperformed competitors across both text-to-video and image-to-video tasks. In text-to-video, it achieved win-loss ratios of 285%, 212% and 160% compared with Seedance 1.0 mini, Veo3-fast and Seedance 1.0, respectively. In image-to-video, the model secured win-loss ratios of 208%, 289% and 164% against the same respective benchmarks.
Kling AI 2.5 Turbo Video Model Text-to-Video Win-Loss Ratio Evaluation Results
Kling AI 2.5 Turbo Video Model Image-to-Video Win-Loss Ratio Evaluation Results
Enhanced Performance with Greater Affordability
The new model’s overall performance showed significant improvements across multiple dimensions, including prompt adherence, motion performance, style consistency and aesthetic quality.
In prompt adherence, the new model further elevated prompt comprehension capabilities, enabling it to handle complex multi-step instructions and causal relationships more comprehensively with greater precision. Users can now more effectively control dynamic video progression, character interactions and scene transitions through descriptive prompts. As a result, generated videos follow temporal logic more closely, render more subtle facial expressions, and convey richer dramatic tension. The combination of these enhancements aligns video output more accurately with creators’ visions and expectations, achieving sharper and more expressive creative storytelling.
Complex Micro-Expression Semantic Adherence with Streamlined Prompts
Kling AI 2.5 Turbo Video Model can generate more dynamic motion and camera movement with greater amplitude, while more accurately simulating real-world physics to produce smoother, more stable visuals. For instance, the new model delivers superior high-motion scenes, like combat and running with camera tracking, as well as complex action scenes, such as figure skating, synchronized swimming and group dance.
Significantly Enhanced Motion Quality with Streamlined Prompts
In terms of style consistency and aesthetic quality, the new model more accurately captures the artistic style of reference images, demonstrating significant improvements in maintaining stylistic consistency. It can faithfully identify elements from reference images, including color, lighting, texture and overall atmosphere, ensuring generated videos more precisely retain the visual characteristics of the source images. The model also achieves greater aesthetic appeal, particularly in lighting and composition, resulting in outputs with greater visual coherence and impact.
Breakthrough Enhancement in Composition and Lighting with Streamlined Prompts
Kling AI 2.5 Turbo Video Model is now available at a significantly reduced price, nearly 30% lower than its 2.1 predecessor. Each five-second 1080p video is now 25 credits, down from 35 credits. This pricing shift makes advanced AI video generation more accessible, empowering creators worldwide to tell their stories with greater freedom and scale.
Deepened Commitment to Empowering Content Creators
This launch reinforces the AI-powered creative platform’s commitment to empowering content creators worldwide, further enhancing the controllability, stability, and consistency of video generation. These improvements lay the groundwork for its broader application in professional creative scenarios such as film and television, short play, gaming, animation, and advertising, while also providing creators with superior video generation solutions. Since its debut in June last year, Kling AI has undergone over 30 iterations focused on enhancing the fundamental quality of its models and introducing more innovative features to meet diverse user needs.
Kling AI also powered New World Is Loading, the world’s first AI-generated series, which has surpassed 200 million views. Kling AI was recently invited to preview its 2.5 Turbo Video Model at the Busan International Film Festival’s Asian Contents & Film Market (ACFM).
Looking ahead, Kling AI will continue to enhance the foundational quality of its models and develop innovative features to build a one-stop creative engine. This will address the diverse needs of more creators, empowering everyone to “tell compelling stories with AI.”
For more information on the Kling AI 2.5 Turbo Video Model, please click here to view the model’s release notes.
About Kuaishou
Kuaishou is a leading content community and social platform in China and globally, committed to becoming the most customer-obsessed company in the world. Kuaishou uses its technological backbone, powered by cutting-edge AI technology, to continuously drive innovation and product enhancements that enrich its service offerings and application scenarios, creating exceptional customer value. Through short videos and live streams on Kuaishou’s platform, users can share their lives, discover goods and services they need and showcase their talent. By partnering closely with content creators and businesses, Kuaishou provides technologies, products, and services that cater to diverse user needs across a broad spectrum of entertainment, online marketing services, e-commerce, local services, gaming, and much more.
Forward-Looking Statements
Certain statements included in this press release, other than statements of historical fact, are forward-looking statements. Forward-looking statements generally can be identified by the use of forward-looking terminology such as “may”, “might”, “can”, “could”, “will”, “would”, “anticipate”, “believe”, “continue”, “estimate”, “expect”, “forecast”, “intend”, “plan”, “seek”, or “timetable”. These forward-looking statements, which are subject to risks, uncertainties, and assumptions, may include our business outlook, estimates of financial performance, forecast business plans, growth strategies and projections of anticipated trends in our industry. These forward-looking statements are based on information currently available to the Group and are stated herein on the basis of the outlook at the time of this press release. They are based on certain expectations, assumptions and premises, many of which are subjective or beyond our control. These forward-looking statements may prove to be incorrect and may not be realized in the future. Underlying these forward-looking statements are a large number of risks and uncertainties. In light of the risks and uncertainties, the inclusion of forward-looking statements in this press release should not be regarded as representations by the Board or the Company that the plans and objectives will be achieved, and investors should not place undue reliance on such statements. Except as required by law, we are not obligated, and we undertake no obligation, to release publicly any revisions to these forward-looking statements that might reflect events or circumstances occurring after the date of this press release or those that might reflect the occurrence of unanticipated events.
For investor and media inquiries, please contact:
Kuaishou Technology
Investor Relations
Email: ir@kuaishou.com
Photos accompanying this announcement are available at
https://www.globenewswire.com/NewsRoom/AttachmentNg/5dbfb860-df3e-4143-a16b-db7cf2d44db1
https://www.globenewswire.com/NewsRoom/AttachmentNg/aea36e3c-ed5f-4597-9567-f09084510fde
https://www.globenewswire.com/NewsRoom/AttachmentNg/279eab00-b279-4ab5-8a2b-53137f8a1e70
https://www.globenewswire.com/NewsRoom/AttachmentNg/3cf8114f-4c30-47bf-8bd3-c75f359085d6
https://www.globenewswire.com/NewsRoom/AttachmentNg/8bfc93a9-fe6e-4836-954a-f922346e2120
